Qwen-MT translation models

POST

/compatible-mode/v1/chat/completions

import os
from openai import OpenAI

client = OpenAI(
  # If you have not configured the environment variable, replace the following line with your API key: api_key="sk-xxx",
  api_key=os.getenv("DASHSCOPE_API_KEY"),
  # Singapore (ap-southeast-1). For another region, replace ap-southeast-1 in the URL.
  # API keys are region-specific. See https://docs.qwencloud.com/developer-guides/getting-started/regions
  base_url="https://ap-southeast-1.maas.qwencloudapi.com/compatible-mode/v1",
)
messages = [
  {
    "role": "user",
    "content": "No me reí después de ver este video"
  }
]
translation_options = {
  "source_lang": "auto",
  "target_lang": "English"
}

completion = client.chat.completions.create(
  model="qwen-mt-plus",
  messages=messages,
  extra_body={
    "translation_options": translation_options
  }
)
print(completion.choices[0].message.content)

{
  "id": "chatcmpl-999a5d8a-f646-4039-968a-167743ae0f22",
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "logprobs": null,
      "message": {
        "content": "I didn't laugh after watching this video.",
        "refusal": null,
        "role": "assistant",
        "annotations": null,
        "audio": null,
        "function_call": null,
        "tool_calls": null
      }
    }
  ],
  "created": 1762346157,
  "model": "qwen-mt-plus",
  "object": "chat.completion",
  "service_tier": null,
  "system_fingerprint": null,
  "usage": {
    "completion_tokens": 9,
    "prompt_tokens": 53,
    "total_tokens": 62,
    "completion_tokens_details": null,
    "prompt_tokens_details": null
  }
}

Authorizations

string

header

required

QwenCloud API Key. Create one in the QwenCloud console.

Body

application/json

enum<string>

required

The model name.

Available options:qwen-mt-plus,qwen-mt-flash,qwen-mt-lite

Example:qwen-mt-plus

object[]

required

An array of messages providing text to translate. Only user messages are supported.

Show child attributes

enum<string>

required

The role of the message. Must be user.

Available options:user

Example:user

string

required

The sentence or text to be translated.

Example:No me reí después de ver este video

object

required

Translation-specific parameters.

Show child attributes

string

required

The full English name of the source language (e.g., Spanish, Chinese, Japanese). Set to auto for automatic language detection. See Supported languages for the full list.

Example:auto

string

required

The full English name of the target language (e.g., English, Chinese, French). See Supported languages for the full list.

Example:English

object[]

Term intervention entries. Each entry maps a source-language term to its desired target-language translation, ensuring consistent terminology.

Show child attributes

string

required

The term in the source language.

Example:grafeno

string

required

The desired translation of the term in the target language.

Example:graphene

object[]

Translation memory entries. Each entry pairs a source-language sentence with its reference translation, guiding the model with prior translations.

Show child attributes

string

required

The statement in the source language.

Example:Puede utilizar uno de los siguientes métodos para consultar la versión del motor de un clúster:

string

required

The reference translation in the target language.

Example:You can use one of the following methods to query the engine version of a cluster:

string

A domain prompt that describes the subject area or style of the text. Must be in English. Example: "The sentence is from Ali Cloud IT domain. It mainly involves computer-related software development and usage methods."

Example:The sentence is from Ali Cloud IT domain. It mainly involves computer-related software development and usage methods, including many terms related to computer software and hardware. Pay attention to professional troubleshooting terminologies and sentence patterns when translating. Translate into this IT domain style.

boolean

defaultfalse

Whether to return the response in streaming mode. When true, the response is returned as server-sent event chunks. qwen-mt-flash and qwen-mt-lite return incremental output; qwen-mt-plus and qwen-mt-flash return non-incremental output (each chunk contains the full sequence so far).

object

Configuration for streaming output. Only takes effect when stream is true.

Show child attributes

boolean

defaultfalse

Whether to include token usage information in the last data chunk.

integer

Maximum number of tokens to generate. If the output exceeds this limit, it is truncated. Defaults to the model's maximum output length.

integer

Random number seed for reproducible results. Range: [0, 2^31-1].

number

default0.65

Sampling temperature controlling output diversity. Higher values produce more diverse text. Range: [0, 2). Only set one of temperature or top_p.

number

default0.8

Nucleus sampling probability threshold. Higher values produce more diverse text. Range: (0, 1.0]. Only set one of temperature or top_p.

integer

default1

Size of the candidate token set for sampling. Larger values increase randomness. If None or greater than 100, only top_p applies. Must be >= 0. Non-standard parameter: use extra_body in Python SDK.

number

default1

Penalty for repeated sequences. Higher values reduce repetition. A value of 1.0 means no penalty. Must be > 0. Non-standard parameter: use extra_body in Python SDK.

Response

200-application/json

string

Unique request identifier.

object[]

Generated content from the model.

Show child attributes

enum<string>

Why the model stopped generating. stop: output is complete. length: output length limit was reached.

Available options:stop,length

integer

Index of this choice in the choices array.

object | null

Currently fixed to null.

object

The message output by the model.

Show child attributes

string

The translation result.

string | null

Currently fixed to null.

string

Fixed to assistant.

object | null

Currently fixed to null.

object | null

Currently fixed to null.

object | null

Currently fixed to null.

unknown[] | null

Currently fixed to null.

integer

UNIX timestamp when the request was created.

string

The model used for the request.

string

Always chat.completion.

string | null

Currently fixed to null.

string | null

Currently fixed to null.

object

Token consumption information.

Show child attributes

integer

Number of tokens in the model output.

integer

Number of tokens in the input.

integer

Total tokens consumed (prompt_tokens + completion_tokens).

object | null

Currently fixed to null.

object | null

Currently fixed to null.